Science of the Total Environment - StopUMTS productie en vuile... · 2018. 5. 24. · 42-Hz square...

Science of the Total Environment 447 (2013) 500–514

Contents lists available at SciVerse ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

Relationship of electric power quality to milk production of dairy herds — Field studywith literature review☆

Donald Hillman a,⁎, Dave Stetzer b, Martin Graham c, Charles L. Goeke d, Kurt E. Mathson e,Harold H. VanHorn f, Charles J. Wilcox g

a Department of Animal Science, Michigan State University, East Lansing, MI 48824, United Statesb Stetzer Electric, Inc., Blair, WI 54616, United Statesc Department of Computer Science and Electrical Engineering, University of California, Berkeley, CA 94720-1770, United Statesd Goeke Enterprises, Mason, MI 48854, United Statese EIT Rockwell Automation, Mequon, WI, United Statesf Department of Animal Science, University of Florida, Gainesville, FL, United Statesg Department of Animal Science, Geneticist, University of Florida, Gainesville, FL, United States

H I G H L I G H T S

► Dairy cows were sensitive to earth currents from neutral-to-ground circuit outlets.► Clamp-on ammeters on grounded-Y down grounds give quick current readings.► Harmonic distorted voltage affects cows' behavior, health, and milk production.► Peak-to-peak current must be measured for full impact of current on production.► IEEE standards should include harmonic current effects on human and animal health.

☆ There are no conflicts of interest. No funding has bework.⁎ Corresponding author at: Michigan State University, 7

MI 48823, United States. Tel.: +1 517 351 9561.E-mail addresses: [email protected], [email protected]

0048-9697/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.scitotenv.2012.12.089

a b s t r a c t
a r t i c l e i n f o
Article history:Received 8 November 2012Received in revised form 21 December 2012Accepted 22 December 2012Available online xxxx

Keywords:TransientsHarmonicsEMF voltage p-pPower qualityMilk productionDairy farms

Public Utility Commissions (PUC) in several states adopted 0.5 volt rms (root mean squared) or 1.0 milliam-pere as the actionable limit for utilities to respond to complaints of uncontrolled voltage. This study clearlyshows that the actionable level should be reduced to 10 mV p-p (peak-to-peak), which is 140 times lessthan the current standard. Dairy farmer complaints that animal behavior and milk production were affectedby electrical shocks below adopted standards were investigated on 12 farms inWisconsin, Michigan, andMin-nesota. Milk production per cow was determined from daily tank-weight pickup and number of cows milked.Number of transient events, transients, voltage p-p, waveform phase angle degree, sags, and sag-Vrms weremeasured from event recorders plugged into milk house wall outlets. Data from 1705 cows and 939 datapoints were analyzed by multiherd least-squares multiple regression and SAS-ANOVA statistical programs.In five herds for 517 days, milk/cow/day decreased−0.0281 kg/transient event as transient events increasedfrom 0 to 122/day (Pb0.02). Negative effects on milk/cow/day from event recorder measurements weresignificant for eight independent electrical variables. Step-potential voltage and frequency of earth currentswere measured by oscilloscope from metal plates grouted into the floor of milking stalls. Milk decreased asnumber of 3rd, 5th, 7th, 21st, 28th, and 42nd harmonics and the sum of triplen harmonics (3rd, 9th, 15th,21st, 27th, 33rd, and 39th) increased/day (Pb0.003). Event recorder transient events were positively correlat-ed with oscilloscope average V p-p event readings. Steps/min counted from videotapes of a dancing cow withno contact tometal in the barnyardwere correlatedwith non-sinusoidal 8.1 to 14.6 mV p-p impulses recordedby oscilloscope for 5 min from EKG patches on legs. PUC standards and use of 500-Ohm resistors in test circuitsunderestimate effects of non-sinusoidal, higher frequency voltage/current common on rural power lines.

© 2013 Elsevier B.V. All rights reserved.

en received for any part of this

50 Berkshire Lane, East Lansing,

(D. Hillman).

rights reserved.

1. Introduction

Uncontrolled electric current injected into the earth in a Grounded-Wye Distribution System (commonly called “Stray Voltage”), NEV(neutral-to-earth voltage), N-GV (neutral-to-ground voltage), or tinglevoltage has been the subject of controversy between dairy farmers,

http://dx.doi.org/10.1016/j.scitotenv.2012.12.089

mailto:[email protected]


http://dx.doi.org/10.1016/j.scitotenv.2012.12.089

http://www.sciencedirect.com/science/journal/00489697

501D. Hillman et al. / Science of the Total Environment 447 (2013) 500–514

some swine and dog kennel operators (Marks et al., 1995), and elec-tric utilities in North America since 1970. Craine (1969, 1982) andCraine et al. (1970) found electrical currents on domestic water sys-tems from primary neutral down-grounds. Jersey cows decreased inmilk production, and cattle decreased water consumption when ex-posed to similar voltages on watering troughs. Some 1300 herdowners filed complaints of electrical interference to the MPSC andAttorneys General of Michigan prior to initiation of AG v ConsumersEnergy MPSC Case No. 11684 in 1998.

A Review of the Problems Associated with Stray Voltage on DairyFarms was published in the Bovine Practitioner (Zdrojewski andDavidson, 1981) and a review of “Sources of Stray Voltage and Effecton Cow Health and Performance” was published in the Journal ofDairy Science (Appleman and Gustafson, 1985). The opinions of“stray voltage experts,” based on limited studies of 60-Hz (Hertz) AC(alternating current), sinusoidal voltage, were published in USDA-ARSPublication 696 (1991), Effects of Electrical Voltage/Current on FarmAnimals: How to Detect and Remedy Problems.

USDA-ARS Publication 696 (1991), called the Redbook, becamethe standard for cow-contact stray voltage adopted by public utilitycommissions and utilities in several states. The standard usuallyaccepted was a minimum of 0.5 Vrms (volt root mean squared) or1 mA (milliampere) of 60-Hz, steady-state voltage, contributed bythe utility, an amount that must be present at cow-contact pointsfor the utility to be responsible for correcting an electrical problem.Cow contact was defined as touching metal water bowls, pipelines,stanchions, stall dividers, and feeding equipment. Power companystray voltage experts use a 500-Ohm resistor in the voltmeter circuit.The theory was that a voltage must be strong enough for the currentto pass through the resistor to affect cows; and if cows do not exhibitphysical signs of electric shock, electricity has no harmful effect. Volt-ages less than approximately 0.5 V, or 1.0 mA current, were consid-ered “not significant” when resistors were in the voltmeter circuitand Wisconsin or Michigan PSC protocol were followed. However,the Redbook contained no information about effects of transients(electrical surges) or harmonics, integer (whole number) multiplesof 60 Hz in North America and 50 Hz in Europe and Asia, generatedwithin circuits and power lines by transients, oscillating at frequen-cies other than 60 cycles per second on power lines. Harmonics,often called electrical noise, may produce humming, buzzing, and rf(radio frequency) radio noise heard near electrical power lines.

Professor Lloyd B. Craine, co-author of the Redbook, acknowledged,“…When consumer equipment consisted primarily of lights, motors,and tube-type electronic equipment, and electrical loadswere relativelysmall, neutral-to-earth voltages and transients were not great prob-lems, due to low neutral currents and the tolerance of the equipment.With increasing use of low-signal-level solid-state computers andmicroprocessors, increasing electrification and automation of farms,and increased loads on distribution lines, the issue of power qualityand tolerable neutral-to-earth voltage is increasingly important.” Crainerecommended, “Transient-effects research is necessary to fully evaluatepower systemeffects on animals” (USDA, 1991, sec 6, pp. 2–4). The pur-pose of this investigationwas to determine if electric power quality andstray voltage were related to changes in milk production of dairy cows.

A dairy company farm-service agent asked a local industrial elec-trician to “look into” farmers' complaints that their cows were affect-ed by stray voltage when utility stray voltage experts said no voltagewas present and the problems were all caused by poor farm wiringand management. Tests were conducted on more than 100 farms inWisconsin, Michigan, and Minnesota. Power quality was measured interms of compliance with defined voltage, frequency, phase generationand current phase delivery efficiency, number and magnitude of tran-sients, harmonics, sags, surges, and outages. Inferior quality power isknown as “dirty electricity” in the electrical industry (Kennedy, 2000;Mazur, 1999). Effects of transients and harmonics in stray voltage ondairy cattle and other farm animals were not previously reported in

animal science literature in our search of the journals. However, electri-cal interference, assumed to be 50–60 Hz “Stray Voltage,” was in theRedbook and ASAE Symposia 1984 and 2003.

2. Materials and methods

Milk and electrical measurements were studied on 11 farms, and legmovements and electrical data from a 12th farm (Table 1). These farmswere selected because of suspected electrical problems that farmersbelieved may have influenced animal behavior and performance.

2.1. Data recording

Data recording equipment were located in the office or milk-roomadjacent to the milking barn or parlor as in Fig. 1.

Transient information was recorded with a FLUKE® Voltage EventRecorder VR101 employing EventView™ Software. The event recorderwas plugged into an electrical outlet in the milk house or in themilking parlor. Time (day, h, min, s), number of H-N (Hot-to-Neutral)and N-G (Neutral-to-Ground) transient events, total number transientoscillations per event, V p-p (voltage peak-to-peak), H-N sags Vrms(voltage root mean square), H-N swells, and wave angle degreeswere recorded by the event recorder. The event recorder accumulated4000 events before it was full and had to be downloaded to computer.

Step-potential voltage and frequency (Hz) were measured frommetal plates (10×15 cm), 1.5 m apart, grouted into the concretefloor of milking stalls as recommended by science advisors (Hobenet al., 1998). Plates were connected via twisted shielded cable, twistedpair, or THHN building wire leads to a FLUKE® 105B Scopemeter SeriesII (100 MHz recording oscilloscope) employing FlukeView™ SoftwareSW90W on a Dell® Inspiron 7000 (laptop computer). Cattle move-ments were recorded simultaneously with a Sony® Handycam VisionCCD-TRV43 videoHi8 (portable video recorder) for part of the period.Computer output was converted to a video signal via Focus Enhance-ments Tview™ Gold Card (pc-card adapter and software) and mixedwith the video signal of the cattle by way of a Videonics MXPro DigitalVideo Mixer model MX-3000 (audio/video mixer) and recorded on aSony Hi8 Video Cassette Recorder EV-C200 (Hi-8 VCR). Electrical im-pulses and cow movements were recorded simultaneously on video-tape by Stetzer Electric, Inc., and analyzed by Essential Regression©ver. 2.218, 1998, with macros incorporated into Excel by Microsoft®.Composite multi-herd data were analyzed as described in Section 2.3.

A BK Precision 4040 20-MHz sweep/function generator was usedto calibrate the remote monitoring oscilloscope by first injecting a42-Hz square wave 2.3-V signal into a battery powered Tektronix720P scopemeter at the plates, and then injecting the same signalinto the wires that would be connected to the plates for monitoringpurposes. The signal was then verified to be the same on the Fluke os-cilloscope in the remote monitoring location as the Tektronix 720P atthe plates.

Milk production was from daily milk tank weights determined bythe milk-hauler and from milk-check statements. Milk (kilograms=2.2046 pounds) were divided by the total number of cows that con-tributed to the tank load. Cows that were too fresh to enter theirmilk in the tank or were receiving medical treatment were not partof the milk herd. In two herds where milk was picked up on alternatedays, weights were handled accordingly to determine average milkper cow per day corresponding to electrical measurements for the2-day period and were analyzed separately.

Cow leg movements (lifting feet, stepping, kicking) of a cow, tiedonly by a rope in the barnyard of herd number 12 (Table 1) locatednear a large substation, were recorded on videotape while electricalactivity on the cow's legs was recorded by oscilloscope. Channel Aleads were attached to EKG patches (electrocardiogram electrodes,3M Red Dot™) placed on shaved skin over the right front (RF, meta-carpal) and right rear (RR, metatarsal, or cannons) and were held in

Table 1Data source: Farm records, Fluke® Event Recorder Transient Deviation Thresholds, oscilloscope, RPM, distribution system, and connections.

Herd I.D. Monitor dates Description Cows milkedavg. no./day

Milk per cowaverage kg/day

Data points weightperiods no.

Recorder transientthresholdVp-p H-N & N-G

1. Kru WI 7/15/99–3/22/00 Event recorder in 110-V outlet in milk room. 110 22.50 243 100 & 502. H-M WI 1/30/00–4/04/00 Tie stall barn, event recorder in milk room. 87 30.49 43 100 & 503. Eri WI 8/1/99–12/20/99 Free stalls, double-12 parlor, event recorder in parlor.

Ground currents from plates imbedded in concrete inmilking parlor recorded by oscilloscope.

366 32.80 13674

200 &100and 100 &50

4. Bey WI 12/17/99–1/19/00 Free stall barn, some tie-stalls. Milking parlor. Eventrecorder in 110-V outlet in milk room. Oscilloscope.

80 27.50 3452

150 & 50100 & 50

5. Pla MN 2/6/00–4/10/00 Free stall barn, double-4 parlor. Event recorder andoscilloscope. Farm located in return path betweensub-station and nonlinear loads. Earth primary neutralreturn conductor.

56 29.50 135 100 & 50

6. Ram MI 6/26/99–7/22/99 Freestalls, double-8 parlor; event recorder; 2-day pickup.Loss of milk production, 56 cows died in previous 2 years,herd dispersed.

110 22.70 13 200 & 100

7. Bel MI 3/5/00–8/01/00 Freestalls, double-6 parlor. Event recorder and oscilloscope;2-day pickup, data for subperiods of time. Loss of milk, cows,displaced abomasums, herd dispersed.

96 31.40 37 100 & 100

8. Wal MI 4/22/00–12/31/00 Oscilloscope, free stall, 2-12 parlor. Some event meter data.Milk loss, cow health. System changed to 3-phase afterinvestigation.

325 31.66 204 100 & 100

9. Gut WI 8/04/00–9/04/00 RPM—Split single-phase primary neutral. 68 27.71 32 NA10. Mic WI 6/10/00–9/04/00 RPM—3-Phase Wye primary neutral. 374 27.94 60 NA11. Mut WI 5/19/00–3/23/00 RPM—3-Phase Delta (floating and grounded)

Primary neutral-to-ground.40 25.26 73 NA

12. Jon MI 7/23/99 Oscilloscope and video camera, substationb200 M., transmission lines near farmstead.Loss of milk, cows and herd.

Average 154 28.13 105Sum 1705 939

502 D. Hillman et al. / Science of the Total Environment 447 (2013) 500–514

place by wrapping athletic bandages. Channel B leads were placed onshaved skin over left-rear (LR) and right-rear (RR) cannons of thecow's legs. Leg movements or steps per minute were counted fromthe videotape and regressed on V per minute (V p-p) simultaneouslyrecorded (once/s) for five minutes by oscilloscope. Movement ofcows while in the milking stall, and shimmering of skin and musclesof cows, heifers in the feedlot, and horses at pasture were recordedon videotape by Stetzer while corresponding electrical impulsesrecorded from the step-potential electrodes appeared on the oscillo-scope and television screen (Stetzer, 1999).

Time (day, h, min, s), peak-to-peak voltage plots, frequency plots,waveform capture (snapshot of waveform required for harmonic spec-trum analysis), and harmonic spectrum analysis (fundamental frequen-cy, total harmonic distortion)weremonitored by oscilloscope, recordedonce per second, and evaluated on FlukeView™.

The oscilloscope recorded once per second (86,400 observationsper day), generating a large amount of data for processing into

Event Recorder Oscilloscope

Fig. 1. Electrical instruments used to record power quality data. The event recorder isplugged into a 120-V outlet in the office and the oscilloscope is connected by twisted,shielded cable to metal plates grouted into the floor of milking stalls.

meaningful values. A computer software program was devised to ex-pedite summarizing event meter and oscilloscope recordings to dailynumbers for analyses.

The Reliable Power Meter, Multi-port Permanent RPM Recorder,Model 1942, and similar RPM recorders were used to monitor powerquality in a trial test at three dairy farms in Wisconsin. The meterwas installed at the interface of the primary and secondary circuitsat the service entrance PCC (point of common coupling) according tomanufacturer's instructions (Reliable Power Meters, 400 BlossomHill Road, Los Gatos, CA). Milk per cow was determined from dailytank weights and number of cows milked as described above. Milk/cow/day was regressed on number of transients per day, describedby codes of events, recorded by the RPM in a combined three-herddata set.

Measures of step potential were recorded by battery-powered os-cilloscope while the primary power supply was completely discon-nected from each farm. During this period there were no detectablechanges in spikes on waveforms indicating the inferior power wasfrom off-farm sources.

2.2. Data sets

Several data sets were constructed to obtain as many farms anddays as possible that included the same measured observations. DataSet 1 consisted of five herds (Table 1, ID 1, 2, 3, 4, 5) that included517 observations of milk production and the number of transientevents per day. Transient events were composed of mixed H-N tran-sient threshold settings at 100, 150, and 200 V in three herds, withcorresponding N-G thresholds set at 50, 50, and 100 V, respectively,while in the other two herds thresholds were 100 V H-N and 50 V N-G.

Data Set 2 was a five-herd data set that included transient eventobservations for 515 days. Independent variables include the sameherds as in Data Set 1, but with Fluke® Event Recorder threshold set-ting in all herds, H-N 100 V and N-G 50 V p-p and included only dayswhen the recorder operated at least 23 h. Independent variables from


EventView™ software were added to the database. They included:number of transient events, number of H-N and N-G transient eventsand corresponding voltages (V p-p), number of transients (oscillations/event), waveform degree angle of the transient event, number of sags(5% below nominal voltage), sag voltage rms (root mean square as in60 Hz), and number of surges/day.

Data Set 3 containsmilk production for four herds (Table 1, I.D. 3, 4,5, 8), 535 data points, and corresponding step-potential voltage andfrequencies (Hz) obtained by oscilloscope from the floor of milkingstalls. Data Set 4 contains milk production from three herds, 165 datapoints, and corresponding step-potential event recorder and oscillo-scope data. In Data Set 5, milk production/cow/day of three herds,165 data points, was regressed on the number of transient eventsmeasured by the RPM at the point of common coupling (PCC) as elec-trical power entered the farm.

2.3. Statistical analysis

Statistical analysis of Data Set 1 was by multiple regression, multi-herd least squares analysis of data with unequal subclass numbers(Harvey, 1990). Dependent variable, average daily milk yield, was ad-justed for Farm (class variable), Date (cubic), Farm × Date Interaction,Number of Transient Events (continuous independent variable), andFarm × Number of Transient Events.

For Data Sets 2, 3 and 4, a series of preliminary statistical analyseswere performed to develop the final mathematical model. The com-puter program utilized was SAS, Inc. (SAS, 1985). Data first werescreened by visual observation and by use of SAS PROC CORR. Severalpotential independent variables were perfectly correlated becausethey were derived from one or more other variables, e.g., some totalsof the number of events or voltages were perfectly or very highly cor-related with averages. These were deleted from further consideration.In least squares ANOVA, farm, date on experiment, and their interac-tions were found to have significant effects (P=.05) on milk yield.To reduce the set of up to 77 recorded or derived electrical measuresto a manageable number, SAS PROC REG (selection = backward) wasperformed on the residuals resulting from the least squares ANOVA(SAS PROC GLM) which included farm, days on experiment (to thecubic order of regression), and their interactions. To be included inthe final analysis, a probability level of Pb0.10 was required arbitrari-ly. The final mathematical model was analyzed with SAS PROC GLMand included farm, date on experiment (cubic), their interactions, aset of electrical variables selected from PROC REG and their interac-tions with farm. All effects were considered to be fixed, except resid-ual, considered to be random.

Fig. 2. Measures of power quality displayed fro

3. Results

Power quality measures were recorded by Fluke® Event Re-corder downloaded to computer for processing and displayed usingEventView™ software as in Fig. 2.

3.1. Transient/harmonic effects on milk production

The number of transient events averaged 14.3±21.7 (range 0 to99 events/day) in Data Set 1. Milk production of cows in five herdsin Data Set 1 averaged 27.08±5.1 kg/cow/day for 517 days (datapoints), and ranged from minimum 14.7 to maximum 35.8 kg for in-dividual farms. Number of cows averaged 165.7±123, ranging from49 to 394 cows milked per day per farm. Difference between farmswas the most significant factor in the mathematical model that wasassociated with variation in daily milk production. Date (cubic) andtransient threshold settings of the event recorder were also significant.Milk/cow/day was negatively related to number of transient events perday (TEV), regression coefficient=−0.0281 kg milk/transient event,significantly linear (P=0.02).

3.2. Transient/harmonic effects on milk production using differentindependent variables

In a second model with milk/cow/day as the dependent variableand farm, sequential dates, threshold settings, and number of tran-sient events as independent variables, the regression coefficient was−0.0287 kg milk/cow/transient event/day, linear (Pb0.001) usingEssential Regression© ver. 2.218, 1998, on Microsoft® Excel software.Transient events accounted for average −0.41±0.62 kg to −2.87 kgmilk/cow/day from average 14±22 and maximum 99 transientevents/day respectively, as measured by event recorder in Data Set1 and illustrated in Fig. 3. Transients are unwanted, short-durationvoltages, called spikes or surges, caused by the sudden release of storedenergy on an electrical circuit.

3.3. Transient/harmonic events v milk/cow on five farms

In Data Set 2, transient events averaged 20±25.9 (standard devi-ation) and ranged from 0 to 122/day on five farms, 515 data points(days), when the recorder was operating at least 23 h per day andthreshold settings were 100 V H-N and 50 V N-G. Eight electrical var-iables were found to be significantly related to milk production as inTable 2. They were (1) total number of transients, (2) total transientdegree angle/100, (3) number of hot-to-neutral transient events,(4) total H-N waveform phase degree angle/100, (5) total sum of

m Fluke EventView™ Computer Software.

0 20 40 60 80 100Number Transient Events

10

15

20

25

30

35

40

Regr. Line

Actual

Milk/Cow = 27.1 kg - 0 .0287 kg/Transient Event/Day

Relationship of Transient Events to Milk/Cow/DayFive Herds, 517 Data Points, Data Set 1

Milk

/Cow

/Day

, kg.

Fig. 3. Relationship of transient events to milk/cow/day, five farms, 517 days. The re-gression coefficient is −0.0287 kg milk per transient event (Pb0.001).


N-G transients, (6) total sum of N-G transients with phase degreeangle=>200°, (7) number of sags (voltageb108), and (8) Sag Volt-age rms.

Transient events and number of transients varied enormously fromday to day and farm to farm. Transient degree angle refers to electricdegrees on a 360° waveform scale, where voltage peaking between0 and 180° is positive and between 180 and 360° is negative. Thenormal 60-Hz sinusoidal waveform peaks are at +90° and −270°.The waveforms recorded were distorted and non-sinusoidal. Theydid not conform to the 60-Hz waveform as from the generator. Suchdistortion causes a shift in time of one voltage waveform relative toother voltage waveforms delivered to the point of service and may re-duce operating efficiency of equipment.

3.4. Oscilloscope “step potential” voltage from the floor of milking stalls

Sixty-nine electrical variables were utilized in statistical analyses.The data were divided into four data sets to be able to managethe data and, within each set, the number of variables were reducedby backward stepwise regression to determine those significant atPb0.10. There were 13 variables found to be significant in the sepa-rated data sets. These 13 variables were then combined and reducedin number by backward stepwise regression, yielding seven thatwere significant (Pb0.10) with one included variable highly correlat-ed with one other (0.998). After removing one of the two most highlycorrelated variables, including the remaining 6 electrical variables(linear) in the model that included farm and date (cubic); farm xdate reduced variation 8.5%. Those six variables were as in Table 3:(1) number of V p-p Readings, (2) Number V p-p Events, (3) NumberVp-p Event Readings, (4) Number of 3rd Harmonics (NH3), (5) Num-ber of 5th Harmonics (NH5), and (6) Number 7th Harmonics (NH7).

Table 2Event recorder measures of power quality for five herds combined, 515 data points, Data S

Total transients(oscillations)no./day

H-N Trans-sients(oscillations)no./day

Total H-N transdegr. angle

Totaldegr

Days – 515 515 385 385 515Mean/day 182 11 −2367 3419Std. dev. 253 17 3349 4618Min. 0 0 −21840 0Max. 1939 89 1630 20,48Ave./trans. −128 297

Yield milk ƒ X, least squares analysis of variance (17) and essential regression by MicrosoftMilk, kg Coef, ƒ X −0.0025 −0.086 −0.014 0.000P Linear>0 b0.001 0.031 0.0018 b0.1

Means = average or sum/day for number days event occurred per possible 515.

3.5. Third, fifth, and seventh harmonic events, and V p-p decreased milkproduction

Variation of harmonic voltage included: (1) number of V p-preadings (No. V p-p Rd), (2) number V p-p events (No. V p-p Events),(3) number V p-p event readings (No. V p-p Ev Rd), (4) numberof 3rd harmonics (N3H), (5) number of 5th harmonics (N5H), and(6) number 7th harmonics (N7H). The most potent single variablein the set of six, and the one that reduced residual variation fromthe base model the most (2.8%), was NH5. Using the linear regressioncoefficient−0.000165withmeanNH5 of 536±980 harmonics/day sug-gests milk yield was reduced −0.15±0.28 kg/cow/day at mean NH5.However, maximum NH5 was 6244 impulses per day×−0.000165=−1.75 kg milk/cow/day maximum accounted for by the 5th har-monic regression coefficient in this data set. The 5th and 7th har-monics combined averaged 753±1348 impulses with maximum of12,065×−0.000365 kg milk/harmonic, thus decreasing milk aver-age −0.27 to −4.4 kg maximum milk/cow/day. Similarly, milk pro-duction decreased as the number of triplen harmonics increased perday. As the sum of triplen harmonics increased from average of 3648impulses to maximum 30,288×−0.000034 kgmilk/triplen, milk de-creased−0.123 to−1.02 kg/cow/day (Pb0.003). Triplen harmonicsare the 3rd harmonic and odd numbered multiples of the 3rd har-monic. The 3rd, 9th, 15th, 21st, 27th, 33rd, and 39th were recordedin this data set with significantly negative effect on milk production/cow/day.

The number of 1st harmonics, 60±30 Hz/day (26,852±30,289/day)was not correlated with changes in milk production/cow in data set 3,apparently because of the large standard deviation in this data set.

3.6. 42nd Harmonic decreased milk production

Data Set 4 consisted of milk production, event recorder and oscillo-scope data, both recording at least 23 h/day, for three herds (Table 1, ID3, 4, 5), 165 data points. Seventy-seven variables were reduced to sixsignificant variables as in Table 4. Including these six variables (linear)in the basic model that included farm and date (cubic), farm × date re-duced residual variation 16.3%. The most potent single variable basedon the analysis including all six variableswas number of 42nd harmonic(NH42=2520 Hz). Selecting NH42 and including it as a single variablein base model reduced residual 4.2%. With linear regression coefficientof −0.007 kg milk/42nd harmonic and mean NH42 of 16±56, appar-ent reduction in milk yield at average NH42 was −0.122±0.39 kgmilk/cow/day with a maximum of 294 voltage impulses/day at2520-Hz frequency accounting for −2.06 kg milk/day. All harmonicswere recorded by the oscilloscope as step-potential voltage and fre-quencies from metal plates in the floor of the milking stall.

Since transients are not produced by the generator, but ratherfrom switching and electronic devices, the non-sinusoidal distortionsof the waveform are related to the H-N and N-G voltages recorded

et 2.

trans.. angle

N-G trans.events

Total N-Gtrans. no./day

Sum N-GAngle=>200°total/day

Sagsno./day

SagsVrms

191 191 144 261 2618.7 79.1 −309 25 269816 163 660 29 26360 0 −3310 0 0

0 90 1024 980 166 184219.1 −73.7 108.3

3 −0.062 −0.0014 −0.0012 −0.03 − .00030 b0.001 b0.001 b0.001 0.02 0.024

Table 3Oscilloscope measures of step potential electrical variables affecting milk production, (Pb0.05), four herds, 535 days, Data Set 3.

No. VpRd No. V p-p events No. V p-p EvRd Ave. V p-p Ev. Rd. No. 3rd harmonic No. 5th harmonic No. 7th harmonic

Mean/day 86,280 3441 41,987 0.0628 2082 536 217Std. dev. 410 2574 33,220 0.0398 4246 980 511Min 82,809 0 4736 0.0268 0 0 0Max 86,400 10,678 83,268 0.1516 30,288 6244 9503Milk, kg. Coef.*X −0.000286 −0.00007 −0.000095 Reference −0.000136 −0.0002 −0.00033P-value b0.01 b0.008 b0.001 Voltage b0.001 b0.005 b0.001

Oscilloscope events were 3 standard deviations from the mean of non-event readings after the last previous event.


by the event recorder. N-G impulses averaged −75.7±37.6 V p-p(range 0 to −190) corresponded to H-N impulses, which averaged−113.4±104 V p-p (range +203 to −307 V p-p).

Oscilloscope event average step voltage, 0.040±0.0116 V p-p,increased linearly as the number of event recorder transient eventsincreased daily (Pb0.001). Step-potential voltages above 0.010 V p-p(10 mV p-p), measured from the floor of milking stalls and in barn-yards, affected behavior and milk production of dairy cows in fourherds for 535 days. This concurs with findings of Polk (2001) inWisconsin herds, working with data collected by science advisors tothe Minnesota PUC (1998).

3.7. RPM (Reliable Power Meter), transient events, and milk production

Numbers of transient events recorded by the RPM are in Table 5.Milk/cow/day averaged 26.8±1.75 kg in three herds (Table 1, ID 9,10, 11); 167 cowswas the average numbermilked/day; 165 data points(days). Herds averaged 40, 68, and 374 cowsmilked/day. Milk/cow/daywas regressed on primary neutral transient events (Table 5) recordedfrom the point of common coupling between primary and secondarycircuits. Herd number, sequential dates and number of cows milked/farm/day were significant and included as independent variables.Number of cows reduced residual variance 6%, and transient events(Code 30-2) reduced residual variance 9% from the basic model residu-al. Milk production decreased−0.029 kgmilk/cow/code 30-2 transientevent/day, (Pb0.001), and similarly for other primary neutral eventcodes. The average number of primary phase transient events wassimilar for phases 0, 1, and 2, but only phase 2 waveform eventswere related to changes in milk/cow/day in this RPM data set. Milk/cow/day decreased as total phase transient events (total of 3 phases)increased/day, regression coefficient −0.001 kg milk/transient/cow/day (Pb0.03). Correlation coefficients for primary neutral transientevents and phase-2 transient events ranged from r=0.47 to 0.86. Re-sults obtained with the RPM were comparable to results with EventRecorders and oscilloscopes at different locations. Large variations innumber of events were recorded from day to day and farm to farm.

3.8. Relation of leg movements to 60-Hz electricity

Leg or foot movements (steps/min) and oscilloscope measuresof voltage from EKG patches attached to shaved cannons of a cowwere recorded and reported in ADSA Presentation Paper No. 03-3116(Hillman et al., 2003b). RF (Right front) to RR (right rear) voltages were

Table 4Event recorder and oscilloscope measures of transient and harmonic electrical impulses aff

Ave. qty NG transØ >200°

NG transØ>200° V p-p

Minsag

Days 122/165 66/165 96/1Ave./day 3.75 −75.7 102Std. dev. 4.5 37.6 51.7Min. 0 −190 84Max. 26 0 107Milk, coef. kg×X −0.027 −0.0017 −0.P-value 0.004 0.10 0.01

recorded on Channel A and averaged 13.2±0.49 mV p-p and ranged8.1 to 14.6 mV p-p. RF movements averaged 3.6±2.7 steps/min (range1 to 8), while RR averaged 10.4±5.7 steps/min. The RR leg was a com-mon ground for both channels. Regression coefficients for RF+RR andtotal steps as a function of maximum–minimum mV p-p were positiveand significant (Pb0.04). Leg movements were significantly correlated(r=+0.89) with step-potential 60-Hz voltage. The procedure waspresented in ASAE Presentation Paper No. 03-3116 in Las Vegas, NV;in CSAE-CSGR PresentationNo. 03-505 inMontreal, Quebec, Ca; in a pre-sentation at ADSA in Phoenix, AZ, in 2003, and at the 12th InternationalConference on production diseases in farm animals at Michigan StateUniversity (Hillman et al., 2003a, 2003b, 2003c, 2004).

Fig. 4 indicates the critical value was 0.9 mV p-p differential be-tween Maximum and Minimum mV p-p during the correspondingminute at which legmovements accelerated significantly with voltageon Channel A. Standard deviations for Channel A mV p-p per minute(5 min=300 observations) were also correlated with RF+RR andTotal Steps per correspondingminute, r=0.88, (Pb0.05), in Fig. 5. Ap-parently, measures of dispersion or differences in potential voltagewere more important than average voltage since neither averages norsumswere related to the number of steps perminute on either channel.

Voltages recorded from LR (left rear) to RR leg on Channel B aver-aged 12.04 mVp-p (range 5.6 to 12.4 mVp-p). LR steps averaged 9.8±5.7 (range 1 to 20) and RR steps averaged 10.4±6.9 (range 2 to 18)steps per minute. Steps were not significantly correlated with voltageon Channel B (P=0.12). Total number of leg movements averaged20±8.7 (range 6 to 40) movements per minute. RR steps were corre-lated with Channel A Max–Min mVp-p (Pb0.07). The LR Steps wascorrelated with RR steps, r=0.99, (Pb0.008). Perhaps Channel A volt-ageswere reflected in LR voltages through the RR leg common ground.

During the 22 min the videotape was recording, the Jonseck cowstepped with her RF foot 3.4 times per minute, RR foot 7.2 times,and LR foot 7.0 times per minute. RF was attached to Channel A, LRto Channel B and RR was a common ground for both A and B. Thecow lifted rear feet twice as many times as front feet. The video cam-era was inadvertently shut off for about 1½ min (2:48 and 2:49); thuslimiting the leg count for that period while the oscilloscope operatedfor 429 s recording once per second continuously.

Step-potential voltage from the ground during the period rangedfrom 18.8 to 22 mV p-p. A voltage drop of 38 to 42% occurred be-tween ground surface and leg skin. This finding is within the 25 to45% of voltage from cow contact (stall divider) to floor plates notedby Ludington et al. (1987).

ecting milk production in three dairy herds, 165 data points. Data Set 4.

imumV rms

Number 21stharmonic

Number 42ndharmonic

Step volt eventrd V p-p

65 17/165 17/65 165.6 467 164 0.040

154 56 0.0100 0 0805 294 0.071

0042 −0.0664 −0.0041 −14.70.07 0.03 0.02

Table 5Transient events recorded by the Reliable Power Meter® in three herds.

Code 25-2 trans.events

Code 26-2 transevents

Code 30-2 transevents

Phase A=0trans events

Phase B=1trans events

Phase C=2trans events

Average/day 12.4 5.5 14.0 33.6 32.6 32.5S.D. 22.8 28.0 18.4 87.2 107.8 63.4Minimum 0 0 0 0 0 0Maximum 257 315 139 1051 1357 681Milk, kg/trans event −0.025 −0.020 −0.029 0.0 −0.0008 −0.003P value linear>0 0.018 0.015 b0.001 Not sig. Not sig. 0.07

Code 25 — RMS voltage/current event.Code 26 — voltage/current waveform event.Code 30 — voltage/current transition waveform event.Phases 0, 1, 2 — Primary phase voltage/current waveform event, equivalent to Phases A, B, and C.


Frequency of voltage during recording with oscilloscope rangedfrom 2.7 MHz to overload at 30 MHzwhile the Jonseck cowwas danc-ing to avoid electrical shocks. Oscilloscope snapshots of the waveformindicate the distorted voltage as in Figs. 6 and 7.

3.9. Electric fields – Electricity travels inside and outside of wires

Electric fields were estimated from step-potential voltage (V p-p)recorded by the oscilloscope in milking stalls at each farm, using im-pedance (resistance) reported for cows (Aneshansley et al., 1995;Appleman and Gustafson, 1985; Craine, 1969; Norell et al., 1983). In-tensity of electric fields (E-field V/m) was estimated from humanmodels of the amperage short circuit (Isc), frequencies from 60 Hzto 1 MHz using Chiba's formula with known short-circuit current, fre-quencies, and height of the object as: Isc=5.4 * 10−9 * H2 * E * f/60.Therefore, E field=(V/R)/((5.4 * 10E−9 * H2)/f60 Hz), and Isc = V/R,where V = step potential V p-p, H = Height of Cow (1.4 m), R =Cow Resistance, f/60 = frequency normalized, and 5.4* 10−9=Constant from Chen et al. (1986). Chen et al. found the formula, Isc=0.108 * h2

m * E0 * fMHz, provided reliable results for predicting short-circuit currents induced by high frequency (>1 MHz) electric fields ina human body.

Estimated E-fields ranged from average, 1.29 kV/m on one farm for54 days to maximum 5.55 kV/m on another farm for 204 days as inTable 7. The actual exposure time (days, months, or years) of farmherds was unknown. The highest E-field observed was 29.6 kV/mcausing a cow to dance in the milking stall while 0.165 V p-p, 625-Hzelectric shocks were recorded on oscilloscope three times during onemilking. That was three times the 10 kV/m exposure of cows to 60-HzE-fields in 28-day trials by Burchard et al. (1998, 1999, 2003).

0.9 2.3 2.4 4.6 6.0

Max - Min mVp / minute

0

10

20

30

40

50

Ste

ps

per

Min

ute

RF+RR

Total

Steps v Voltage (Max-Min)/Minute Jonseck Dancing Cow 7/23/99Jonseck Dancing Cow 7/23/99

Fig. 4. Steps/minute increased as maximumminus minimum voltage increased/minuteon the cow's leg. Voltages from the ground were recorded by oscilloscope attached toEKG (electrocardiogram) patches on right front and right rear legs.

THD (Total Harmonic Distortion) is a measure of the percentage ofharmonic voltage >60 Hz relative to the fundamental first harmonicvoltage (Kennedy, 2000). The Institute of Electrical and ElectronicEngineers in publication IEEE 519-1992 (1993) sets current limits onthe utility side of the meter THD as 5% of the fundamental harmonic.THD was outside these limits on the five farms studied as in Table 6.THD on phase wires also exceeded IEEE 519 limits in power qualitystudies conducted by utilities in Indiana (Tran et al., 1996) and Kansas(Li et al., 1990). Similarly, limits for TDD (Total Distortion Demand)which is the end-user contribution to distortion on the utility lineare also recorded in IEEE 519 (1993).

4. Discussion

4.1. Sources of neutral-to-earth voltages and consequences

Electric current flows through conductors, including the earth, theair, water, metal equipment, animals, and man in its return path to itsoriginal source as well as on wires. In a grounded-neutral wye distri-bution system, 65–75% of residual current returns to the earth sincethe neutral wire is inadequate to return the current to the substation.In addition electric and magnetic fields radiate from the conductors.

The finding that milk production decreased as event recorderneutral-to-ground voltages increased from an outlet in a milk-roomcorresponds to other reports. Neutral-to-ground voltages during tran-sient events averaged−137.9 V p-p (range 20 to−160) inmilk-roomoutlets of five farms in this study.

4.2. Step-potential voltage

The pathways, the step-potential voltages from the floor ofmilking-stalls, and the transient events from nonlinear loads in thepresent study were comparable to unbalanced loads (from primary

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

mVp Standard Deviation / Minute

0

5

10

15

20

25

RF

-RR

Ste

ps /

Min

ute

StDev Regr

Actual

RF-RR Steps v mVp Standard DeviationJon-Cow 7/23/99 R = 0.88

Fig. 5. Relationship of steps/minute to standard deviations of mVp/minute. (The stan-dard deviation is a measure of voltage potential relative to average voltage).

Fig. 6. Oscilloscope plot from EKG patches on leg of dancing cow from step voltage off the floor where the cow is standing. Voltage ranges from about 6.4 to 16.1 mVp, 7.25 to25 MHz frequency in this oscilloscope snapshot, Jonseck Farm, 7/23/99.


or secondary circuits) as described by Ludington et al. (1987). “Bond-ing allows current to flow to metal water pipes, lightning protection,and branch circuit equipment ground wires. The amount of currentflowing in each path will depend on the impedance of the path.”Impedance of cows decreases as frequency of voltage/current in-creases, thus cows receive higher energy (amperage) from higherfrequency harmonics than from 60-Hz sinusoidal electrical impulses(Aneshansley et al., 1995; Aneshansley and Czarniecki, 1990); andcows may not feel the higher rf charges as most humans do nothear or feel (perceive) rf waves unless aided by a radio receiver atproper Hz (frequency). An AM radio produces static (audible noise)in the presence of electrical disturbances in the environment ornear power lines and can be used as a simple test instrument. Further,Ludington noted “the service panel is a divider of neutral voltage.[And] the animal would almost always be a current path in serieswith, and in parallel to, other resistances.” In addition, “The stray volt-ages, as measured with a 500-Ohm resistor, are fractions (ca 10%) ofthe applied voltages,” according to the data (Ludington et al., 1987).

4.3. Transient voltage

Transient voltage is a temporary unwanted voltage, caused by thesudden release of stored energy in an electrical circuit. A transientvoltage is produced from stored energy contained in the circuit induc-tance and capacitance. Oscillatory transient voltages are commonly

Fig. 7. Oscilloscope waveform showing the distorted notched n

caused by turning OFF high inductive loads and by switching largeutility power-factor correction capacitors. A phase shift occurs be-tween alternating voltage and current in an inductive circuit. Thegreater the inductance the larger the phase shift, in which currentlags voltage. Because a change in frequency changes inductance reac-tance of a circuit, any change in frequency delivered to the load has aneffect. Because non-linear loads produce harmonics, harmonics havean effect on the power distribution system and loads (Kennedy, 2000,p. 34). The trend of increasing harmonics on power lines (Najdawiet al., 1999) and adverse effects of inferior quality power on cus-tomers' equipment have been reported (De Andrea, 1999; Kennedy,2000; Mazur, 1999).

4.4. Harmonic distortion

Harmonic distortion is caused by non-linear loads in electroniccircuits such as variable speed motor drives (speed depends on fre-quency), electronic ballast used in lighting circuits, switch-modepower supplies in personal computers, printers, and medical testequipment such as MRI (magnetic radiation imaging) and cellular-telephone relay station transmitter neutral wires. Harmonics are es-pecially a problem where there are large numbers of computers andother nonlinear loads that draw current in short impulses. Triplensresult from single-phase nonlinear loads that draw current onlyduring the peak of the voltage waveform. These loads combined in a

on-sinusoidal transients carried on the 60-Hz waveform.

Table 6Steps per minute and mV p-p from shaved skin at cannons of dancing cow. Maximum (-) minimum mV p-p during the corresponding minutes was correlated with steps R=0.89(Pb0.05).

Time Leg/foot moves Channel A: Voltage measuredper minute Shaved skin on leg

Minute LR RF RF+RR All Legs mV p-p MV p-p mV p-p MV p-p mV p-p Max mV p-pSteps Steps Steps Steps Ave Sum St. dev. Min Max Min mV p-p

2:44 10 4 16 26 13.21 806 0.4011 12.3 14.6 2.32:45 11 1 11 22 13.20 792 0.5150 11.9 14.3 2.42:46 1 3 5 6 13.32 799 0.2045 12.9 13.8 0.92:47 7 8 18 25 13.03 782 0.6293 10.0 14.6 4.62:50 20 2 20 40 13.16 790 0.6963 8.1 14.1 6.0Sum: 49 18 70 119 65.9 3969 2.4462 55.2 71.4 16.2Ave 10 3.6 14 24 13.18 793.7 0.4892 11.04 14.28 3.24St dev 6.9 2.7 6 12.1 0.095 8.27 0.1743 1.76 0.3059 1.818Min 1 1 5 6 13.02 781.7 0.2045 8.1 13.8 0.9Max 20 8 20 40 13.32 805.8 0.6963 12.9 14.6 6.0


three-phase circuit produce triplen harmonics. Triplens do not cancelone another but are additive and return exclusively through the neu-tral conductor. The resulting magnitude of the neutral current mayexceed the capacity of the neutral conductor, and since there are nocircuit breakers in the neutral, overheating of circuits occurs andmay cause fires (Kennedy, 2000; Mazur, 1999).

Nonlinear loads include all types of electronic equipment that useswitched-mode power supplies, e.g., rectifiers converting ac to dc, in-verters converting dc to ac, arc welders, battery chargers, and switch-mode AC–DC power supplies that run radio stations and cellular tele-phone relay stations. All these devices change a smooth sinusoidalwave into irregular distorted wave shapes (Kennedy, 2000, p. 51;Mazur, p. 89) as were captured by oscilloscope at all farms in thepresent study (Figs. 6 and 7).

Excessive leg/foot movements, stepping and shiftingweight repre-sent abnormal behavior that can be disturbing during milking. Thesemovements correspond to the samemagnitude of voltage that causeddecreases of milk production in the present study and do not conflictwith other research. Norell et al. (1983) observed that cows pickedup a front foot twenty-two percent of the time under no shock condi-tions and considered these random foot movement. However, thelowest treatment current in their experiments were 1.0 mA, equiva-lent to 0.36 V, considering 360-Ohmmouth to front hooves resistance,determined in the experiment to cause cows to raise a front foot.Lower uncontrolled voltages were not measured. Similarly, Brennanand Gustafson (1986) found “an unexpected level of response tocontrol (0 cycles of 60 cycle AC treatment).” Controls gave escape re-sponses (opening themouth) 26 times per 30-second test period com-pared to treatment cows receiving 1, 3, 8, 15 cycles responding 56 to58 times and cows receiving 30 cycles responding 75 times per 30 s.

In the present study, uncontrolled impulses of various frequencies(Figs. 6 and 7) were recorded by the oscilloscope as the cow was

Table 7Electric fields for average and maximum event voltage (Vp), estimated from Chiba in Chen

Electric-fields and harmonic distortion in milking stalls

Recordeddays

Event (1) ave.±std.deviation

Total harmonicimpulses

Different1st–42nd

Farm No. Volts (Vp-p)/day No./day Number

Eri 115 0.056±0.01 39,805±23695 16Bey 76 0.050±0.01 34,593±10411 7Pla 108 0.032±0.01 9,746±9403 11Bel 54 0.039±0.01 21,553±26442 29Wal 204 0.063±0.04 44,084±33,201 25

(1) Event on the oscilloscope was ±3.0 standard deviations from the mean of voltages follVoltage = average recorded during events for the period.

dancing. Cow resistance to current decreases as frequency increases(Aneshansley et al., 1995), thus the high-frequency, non-sinusoidalimpulses from rural power lines recorded at the farms studied ap-parently produce different responses than 60-Hz, steady-state, sinu-soidal currents used in experiment station trials reported in journalarticles.

The present observation of leg movements in relation to voltagedifferential was similar to the conclusion by Norell et al that 13.8% ofcows that were trained to perform an avoidance response, respondedto 1.0 mA 60 Hz AC from mouth to rear hooves (Appleman andGustafson, 1985; Norell et al., 1983). Milk productionwas not reportedfor cows responding to various voltages/currents in the sensitivity ex-periments (Appleman and Gustafson, 1985; Norell et al., 1983; StrayVoltage Symposium, 1984; USDA, 1991). The assumption that sensitiv-ity or perception of electricity by cows equates to effect onmilk produc-tion and animal health has not been tested sufficiently to permit a validconclusion.

Milk production decreased 11 to 17% compared to controls whencows were exposed to 5.0 mA intermittent 60-Hz, steady-state elec-tric shock (Lefcourt et al., 1982). Reliable laboratory experiments de-termining effects of electrical power quality on milk production andhealth of dairy cattle comparable to those found on rural utility lineshave not been reported. All of the previously reported experimentshave been conducted with 60-Hz, steady-state sinusoidal currents. Ingeneral, stray voltage experiments (Stray Voltage Symposium, 1984;USDA, 1991) have jeopardized the probability of finding statisticallysignificant effects of electricity on milk production because numberof cows/treatment were inadequate, sources of variance within andbetween groups were too large and often not considered, electricalexposure time of cows too limited, and turnover of cows too excessiveto provide reliable data and valid conclusions (Behr, 1997). Those lab-oratory exercises have very little in common with farm experiences

et al. (2000), and THD (total harmonic distortion) percent of 1st harmonic voltage.

harmonics E-field ave. E-field max Voltage harmonic distortion %

kV/m ave. KV/m max THD ave. % THD max. %

2.678 3.740 67.8 132.13.585 4.301 19.3 70.92.150 4.482 29.9 79.61.293 3.430 22.7 75.02.298 5.551 90.3 23.6

owing the last event.


where exposure may be lifetime, constant, variable in magnitude andfrequency, and sunken resources are limited.

Increases of primary neutral current flow into the grounded neu-tral network (IPA) by primary neutral current from a neighboringfarm and primary neutral current from on-farm loads was describedby Gustafson and Cloud (1982). Secondary neutral current due to un-balanced on-farm loads, interconnection of equipment and circuitneutral conductors, wiring faults, poor connectors, improper use ofneutral conductors, and tree branches brushing power distributionlines may each contribute to increased neutral-to-earth transientand harmonic currents in the livestock environment.

Results of the present study confirm a 1980 report that low-levelneutral-to-ground voltages and transients are a significant problemin some milking areas. And, jumpiness, kicking, refusal of some cowsto enter the milking parlor, and reduced milk production are somemanifestations of the problem (Gruesenmeyer, 1980).

Polk (2001), one of the science advisors to the Minnesota PublicUtilities Commission, using the science advisor's data (Hoben et al.,1998) shows considerable scatter of milk/cow/day when step voltagewas less than 0.01 V (10 mV), but a linear relationship between milk/cow/day and step voltage above 9 mV. He noted that when one con-siders only the V-lu (voltage at low electric use) above 9 mV the cor-relation coefficient betweenmilk/cow/day and V-lu becomes+0.994.However, because of the small number of farms, 3 over 9 mV, theP-value becomes 0.069. Also, on three farms where step voltage was9 mV or larger, the value of milk/cow/day decreased linearly withsoil resistivity to current. Results of the present study, when numberof transients, frequency of impulses, and the number of harmonicsper day are considered, support the position of Polk that 10 mVpeak to peak is a critical voltage level on some farms.

Exposure of cows under controlled laboratory conditions to a10 kV/m, 60-Hz electric field and a uniform horizontal magneticfield of 30 μT (microtesla) for 28 days, has shown physiological effectsthat are potentially adverse (Burchard et al., 1998, 1999). Burchard etal. (1998) found a significant increase in quinolinic acid and a trend to-wards an increase in tryptophan in cerebrospinal fluid consistent witha weakening of the blood-brain barrier due to exposure to the electricand magnetic fields. Burchard et al. (1999) also found decreased con-centrations of magnesium and increased concentrations of calciumand phosphorus in blood plasma and decreased concentrations ofiron and manganese in cerebrospinal fluid of dairy cows and heifersexposed to 60 Hz, 10 kV/m electric fields, and 30 μT magnetic fields.Milk decreased 5.97%, fat-corrected milk decreased 13.78%, milk fatyield decreased 16.39%, while feed dry matter intake increased 4.75%during 28-day reversal trials with 16 mid-lactation cows at McGillUniversity, Quebec, Ca (Burchard et al., 1996). Groups were alternated:1st period current Off-On-Off, 2nd period On-Off-On for each 28-dayperiod so cows were exposed for 84 days, total. Results were differentthan a previous trial during which milk fat was higher from exposedcows compared to unexposed. The large difference in fat secreted inmilk by exposed cows has not been explained (Burchard et al., 1996).However, fat secretion in milk from electrically charged cows waslower than from unexposed controls in four of five experiment-stationreports of effects of electricity on dairy cows (Gorewit et al., 1992;Aneshansley et al., 1992).

Observations in the present study indicate that step-potentialcow-contact current in milking stalls was sufficient to cause cattleto lift their feet to avoid electric shock from the floor or ground, andto decrease milk production without contacting metal, e.g., waterbowls, feeders, or stall pipes. Thus, the assumption that cows are af-fected by electricity because they are repelled from shocks on waterbowls and metal feeders, etc., may be true. But, the present dataalso reveals that ground currents that were conducted, or coupled,through feet and legs without touching any metal objects affected be-havior (stepping) and milk production. This concurs with reports thatrats exposed to 150 V/cm electric fields reduced water consumption,

gained less weight, and had lower levels of cortisol in blood in 9 of 10trials, although exposure was through the air, and water was notconnected to electric circuits (Marino et al., 1977).

In contrast, blood cortisol of cows increased temporarily uponexposure for a few minutes during or near milking (Stray VoltageSymposium, 1984; USDA, 1991). Effects on changes in blood adrenalsteroids over long periods of exposure and awide range of frequenciesas found on farms have not been reported but may cause pituitary–adrenal fatigue as in Addison's disease or other impairment of health.

Recent discovery of the effect of external electric fields on mem-brane harmonic oscillations, caused by ions whose collisions withthe membrane surface influence properties of a single lipid chainmay be key to understanding electrical effects on cattle (Wojczakand Romanowski, 1996). Cows depend on microbial fermentation ofingested feeds to supply acetic, propionic, and butyric acids for ener-gy and for formation of milk fat. Electrochemical effects of modulatedVHF fields on the central nervous system (Bawin et al., 1975) mayhelp explain the significance of cerebrospinal fluid protein and elec-trolyte (calcium ion) modifications by electric fields in dairy cows(Burchard et al., 1998, 1999). Autonomic nervous system response,such as epinephrine reducing blood flow through the udder of cattleunder stress (Appleman and Gustafson, 1985; Lefcourt and Akers,1982; Stray Voltage Symposium, 1984) could explain reduced milkproduction but has not been adequately investigated in relation toelectrical shock (Hillman, 2002).

A review by California Health Services Department prepared forthe PUC, reveals human health risks from electric and magnetic fieldsfrom power lines in the home or workplace (Neutra et al., 2001). Chenet al. (2000) reported that ELF (extremely low frequency, 60 Hz) inhi-bition of differentiation of Friend erythroleukemia cells was dosedependent on electromagnetic exposure; and because ELF inhibitsthe same enzyme in-vitro as phorbal esters, phenobarbital and dioxin,it falls in the same class of carcinogens that proliferate but do notcause cancer. Human colon cancer cells increased six-fold during ex-posure to electromagnetic fields in-vitro (Phillips et al., 1986). Electri-cal exposure disturbed melatonin secretion patterns in blood by thepineal gland (Burch et al., 2000), increased brain cancer and leukemiaamong electrical workers (Loomis and Savitz, 1990; Thomas et al.,1987), increased leukemia in children (Loomis and Savitz, 1990),and decreased T lymphocytes in power plant workers (Nakata et al.,2000) indicating a wide range of physiological pathological conditionshave been related to EMF exposure. A higher rate of suicide amongutility electricians and linemen than utility workers not employed inthose jobs, suggests increased risk of mental depression and disturbedsleep patterns upon chronic exposure to low frequency electromag-netic fields (Van Wijngaarden et al., 2000), and further suggests elec-tric field or electromagnetic field involvement with central nervoussystem functions (Bawin et al., 1975).

4.5. Power quality test meter with true RMS volt peak-peak

Power quality problems such as harmonics, sags, or swells involvedistortion of the sine wave. The correct measurement tool for a powerquality problemmust accurately measure the characteristics of a totaldistorted sine wave (Graham, 2002, 2003, 2006).

In 1994, the Wisconsin SVAT (Stray Voltage Analysis Team), madethe choice that a SVAT investigation would include only V p (not Vp-p) readings (Dasho et al., 1994); and this decision was adopted bythe WI Public Service Commission as well as by Minnesota andMichigan's Public Service Commissions. All three states' utility commis-sions measure with instruments such as the SVM-10 and Waverider,adjusted to read only peak (not peak to peak) values, thus missinghalf of the distorted wave form, giving a false reading. Since theMidwest USA utilities' voltmeters do not read peak-peak values, theyuse “average peak” readings, missing the distorted waveforms, and re-port 25 to 50% below True RMS readings as published in Power Quality


Primer (Kennedy, 2000, pp. 180–184). A FLUKE® 105B ScopemeterSeries II instrumentwas used in our 12-farm study; and it recorded volt-age, amperage, and frequencies of the complete sinewave. According toAneshansley, “The combination of equal amounts of 60 and 180 Hzwithdifferent phase shifts and their lack of sensitivity to DC bias indicatesthat cows are sensitive to peak-to-peak voltages and not peak or rms”(Reinemann et al., 1999).

Many “stray voltage experts” including Public Service Commis-sions, government officials, and utilities may need to update theirmeasurement techniques and knowledge of measuring tools. Use ofa True RMS clamp-on ammeter to measure AC and DC current onthe PN-E (primary neutral-to-earth) down-ground at the transformerpole is a simple method to determine the source and magnitude ofthe grounded-Y (Wye) utility's contribution to primary and second-ary neutrals of the electrical system.

Use of a 500-Ohm resistor in the volt meter test circuit for powerquality effectively eliminates from consideration the electrical powerline harmonics, radiofrequency, and microwave measures that werefound to be harmful in this study andmay givemisleading or unreliableinformation to investigators and herd owners. Studies of the effects ofvarious electrical frequencies and harmonics on animals and humansand the physiological processes affecting behavior, health, reproduc-tion, and productivity deserve further attention.

Resistance on a circuit can be measured with an Ohm meter andneed not depend on inaccurate hypothetical 500-Ohm resistance.Appleman and Gustafson (1985) reported that 94.6% of cows weresensitive to 4 mA or less current. Norell et al. (1983) demonstratedthat for a mouth-to-all-hooves pathway, 10% of cows had a resistanceR=244 Ω (Ohm) and 90% had Resistance=525 Ω. “In this case, 10%of the cattle exposed to 1.0 V mouth-to-all-hooves shock would re-ceive a 4.0 mA or greater shock; whereas 90% of cattle would receivea 1.9 mA or greater shock.” Norell reported that specific avoidance re-sponses were exhibited 13.8% of the time at 1.0 mA of current. Signif-icant increases of response rates occurred for each 1.0 mA incrementcomparison up to 4.0 v 5.0 mA paired test, namely: 2.0 mA=30% re-sponse; 3.0 mA=69.2%; 4 mA=92.3% response, and 5.0 mA=98.4%response (Appleman and Gustafson, 1985, p 1558).

4.6. Subsequent research and related studies

4.6.1. Water drinking reluctance behaviorDairy heifers decreased water consumption 32% when the water

trough was charged with 3.0 volts and reduced water consumption52% when the troughs were charged with 6.0-V, 60-Hz power linecurrent compared to no current (Craine, 1969; Craine et al., 1970).

Cowswere reluctant to drinkwater at all the farmswe tested. Theyexhibited “lappingwith the tongue,” a sign of testing thewater and re-luctance to drink. Sincewater consumption is mandatory formilk pro-duction and good health in animals, it most probably contributed tothe demise of many herds.

The observation that cows were reluctant to drink water on farmsreporting stray voltage and decreased milk production led to our mea-suring current (20–40 mA p-p) in the water on farms reporting strayvoltage in Michigan. We found that milk production decreased as tran-sient, harmonic, and rf (radiofrequency) currents increased, and asstep-potential voltage increased daily. We were not able to find anyNorth American agricultural literature reporting the relevance of rfandMW(microwave frequency) currents to dairy cowbehavior, health,andmilk production prior to our study (Hillman, 2008, 2012; Hillman etal., 2011a, 2011b) and believe more research on this topic is necessary.

Scientists have observed that the fundamental physical composi-tion of water can be changed by weak alternating magnetic fieldsat the cyclotron frequency combined with a weak, static dc field(Zhadin, 2010; Del Giudice and Giuliani, 2010). Similar findings werereported by Abraham R. Liboff, while working at the U.S. Naval Medi-cal Research Center in Bethesda,MD, and later as a physics professor at

Oakland University Rochester, MI (Liboff, 1985; McLeod et al., 1987)and Carl Blackman, at the U.S. Environmental Protection Agency(EPA), Washington D.C. (Blackman et al., 1985). Liboff reported thatthe inorganic nutrients: calcium, potassium, and magnesium becameimmobile in the water in experiments with mice.

Cows that are genetically capable of producing over 100 lb (50+kg)milk daily require about 70–100 grams or more of calcium secreted inmilk daily. If ingested calcium, potassium, andmagnesium are immobilein themetabolic systemduring electromagnetic exposure, Liboff's theorymay explain periparturient hypocalcemia (so-called milk fever), rumenstasis, displaced abomasums, and impaired uterine recovery from infec-tions permitting failed reproduction andmastitis post-calving, as well asdecreased water consumption and milk production.

4.6.2. Water lines frequently carry EMF into homes as well as barnsIn 2004, the Lansing Board of Water and Light, Lansing, MI, found

high levels of electric current entering the Hillman home and installeda dielectric coupling on the waterline to stop the electromagnetic fieldsfrom entering the home (Hillman, 2007).

Similar reports of EMF on water lines have been reported byWertheimer et al. in studies from 1979 to 1995 (Lanera et al., 1997)and by Stetzer (2001) in his video, Beyond Coincidence – The Perilsof Electrical Pollution.

4.6.3. Harmonic distortion on farm power lines and on substationsOur observations that harmonic frequencies generate elevated

levels of current on the neutral wire of a grounded-wye distributionsystem concur with reports of harmonics on utility substations andfarm power lines. Tran et al. (1996), an Engineer of PSI Energy, Inc.,Plainfield, IN, et al., reported that “Triplen harmonics, particularlythe 3rd, add in the neutral and have little diversity between loads.The higher neutral currents may cause significant problems. Neutralto earth voltages will increase near the substations which could in-crease stray voltage complaints. … This paper provides fundamentalunderstanding of triplen harmonic influence on stray voltage andEMF related to multi-grounded wye electric distribution systems.”Tran made reference to USDA Publication 696 for stray voltage prob-lems on animal farms; but USDA Pub. 696 contains no informationabout harmonics nor frequencies other than 60 Hz.

Similar to Tran's findings, Kansas engineers, measured electricpower harmonics, 2nd through 63rd, on five rural substations andseven farms in Kansas, where maximum THDJ ranged from 8.2 to34.2% on farms (Li et al., 1990). Gustafson et al. (1979) in responseto farmer's complaints, recorded 83 harmonics near a DC transmissionline in Minnesota.

4.6.4. Radio-frequency interference on power linesThe coupling of external electromagnetic fields to transmission

lines was described by Albert A. Smith, Jr., a Senior Engineer for IBMCorporation. The effect of induced currents can range from noise oncommunication lines and errors in digital circuits to equipmentdamage and even personnel hazards. Some of the more well-knownsources of electromagnetic fields include nearby lightning strikes,AM, TV and FM broadcast stations; radars; industrial, scientific andmedical (ISM) equipment; automobile ignitions; personnel electro-static discharge; the esoteric nuclear electromagnetic pulse (NEMP);and power supply noise and switching transients inside electronicequipment (Smith, 1989). Smith's book illustrates causes and conse-quences of shielding circuits, spacing of transmission cables, and IEEEreferences to research.

4.6.5. Health and reproduction impaired by EMF exposureCows and other animals exposed to electrical stress over long

periods of time develop an analgesic effect, docile, unresponsive tostress, and may not exhibit a physical reaction to electrical charges.This opioid effect results from accumulation of dopamine in certain


sections of the brain. It is excreted in the urine and has been usedas a marker for electrical stress when other sources of stress arecontrolled (Brown et al., 1991; Buchner and Eger, 2011; Milham andStetzer, in press).

Failed reproduction was a common impediment of dairy herdsafflicted with extraneous electricity. Induction of lymphopenia, a com-mon result of electropathic stress, caused luteal dysfunction in cattle(Alila and Hansel, 1984). Retained CL (corpus luteum) on ovaries is acommon cause of failed estrus in dairy cows (Kristula et al., 1992).Failed conception of experimental cows subjected to electricity wasoften overcome by administration of prostaglandins F2 alpha (Lutalyse)to cows not pregnant by 50-days post-partum in complete lactationelectrical exposure experiments (Gorewit et al., 1992). Lutalyse, whichremoves the CL, causes estrus within 120 hours, and may have biasedexperimental effects of voltage on evidence of estrus and reproductionin some experiments (Shaw and Britt, 2000).

Displaced abomasums and rumenitis associated with poor muscletone in cattle were common on farms with uncontrolled voltage andis comparable to the ulcers and gastro-intestinal pain as recognizedsymptoms of electropathic stress in humans and other animals (Selye,1950, 1951; Rea et al., 1991; Dahmen et al., 2009).

Exposure to weak EMF resulted in deformed embryos and off-spring in laboratory animals (Delgado et al., 1982; Moh'd-Ali et al.,2001) and abnormal chick embryos (Juutilainen et al., 1987). Muta-tions of salmonella microbes exposed to weak 100-Hz fields couldaccount for the more common outbreaks of uncommon diseases andalso raises questions about the effect of EMF on the health of ruminantmicrobial populations. Exposure of cows to low-level EMF resulted inalteration of circadian rhythms and some leukocyte differentiationantigens compared to unexposed cows (Stelletta et al., 2007).

Dairy cows on farms and dogs in commercial breeding-for-researchkennels failed to conceive when induced current from near power lineswas found on the metal cages near Kalamazoo, Michigan (Marks et al.,1995). Exposure to induced current increased length of estrus and pro-gesterone content of blood in cows during 28-day exposure periods(Burchard et al., 2003). Repeated acute stress caused a luteinizing hor-mone surge to be missing during the follicular phase of ovulation indairy heifers (Stoebel and Moberg, 1982). Induction of lymphopeniacaused luteal dysfunction in cattle (Alila and Hansel, 1984). Prolongedstress affects estrous cycles and prolactin secretion in sheep (Przekopet al., 1984).

Likewise, early pregnancy loss andmiscarriage of women and poorquality sperm in men were associated with exposure to magneticfields (Juutilainen et al., 1987; Li et al., 2002, 2010). Chromosomalabnormalities were in lymphocytes of humans exposed to power fre-quencies (Nordenson et al., 1984). Genetic defects occurred in off-spring of power frequency workers (Nordstrom et al., 1983).

Electrical charge and EMF have been shown to proliferate andexacerbate neuroendocrine stress and cortical hormones in blood ofcows (Gorewit et al., 1984a,1984b), in sheep (Przekop et al., 1984),and in humans (Buchner and Eger, 2011; Eskander et al., 2011).

Decreased fibrinogen in blood of cows after three weeks exposureto ground currents was a significant discovery in a Minnesota farmherd (Hartsell et al., 1994). The “low fibrinogen” corresponds toreports of DNA SSB (single-strand breaks) and DNA DSB (doublestrand breaks) in human fibroblast cells of persons exposed to EMF,using the comet assay for DNA (Ivancsits et al., 2002, 2003). Nonther-mal DNA breakage by mobile phone radiation (1800 MHz) in humanfibroblast cells and in transformed GFSH-R17 rat granulosa cells invitro indicates serious deleterious effects of RF-MF radiation (Diemet al., 2005).

“The International Agency for Research on Cancer (IARC) has clas-sified ELF EMF as ‘possibly carcinogenic,’ a classification which neces-sarily implies that the epidemiological link (e.g., EMF and leukemia)may be causal and that directly or indirectly, weak ELF magnetic fieldsmay promote DNA damage; that is, they are genotoxic” (Crumpton

and Collins, 2004). Report of DNA damage by exposure to low-levelEMF corresponds to the blood chemistry reports from cows (Hartsellet al., 1994). In addition, Hartsell et al reported increased lymphocytecount, decrease ofwhite blood cells, decrease in segmentedneutrophils,decrease inmonocyte count, and increased SCC (somatic cell counts) inmilk after cows were exposed to ground current for 17 days.

A relationship of childhood leukemia to 3rd, 5th, and 7th harmonic(180–420 Hz) current in the living environment of childrenwas reported(Kaune et al., 2002; Wertheimer & Leeper, 1979, 1982). Also, childhoodleukemia was 4.3 times higher among children whose bedrooms regis-tered 4 mG, (0.4 μT-microTesla) or higher, the threshold breakpointchosen, compared to those with 1.0 mG (0.1 μT) or less in their bed-rooms in Japan (Kabuto et al., 2006).

Maisch (2010, 2012) has explained the Procrustean Approachused by standards-setting committees and has described irrefutableexperiences of customers suffering from excessive exposure to elec-tromagnetic fields, including radiation through the wall from SmartMeters in Victoria, Australia. Michigan's Public Service Commissionis involved in two smart meter cases (MPSC E-Docket Case #U-16129, 2011, MPSC E-Docket Case # U-17000, 2012).

Since EMF interferes with the autonomic nervous system, controlof the neuro-endocrine system which controls essentially all func-tions of the body toward homeostasis, logically a long list of chronicsymptoms are possible, and not necessarily the same in every person,but largely dependent on the individual DNA tolerance or range for thefunction or dysfunction of a particular organ, tissue, or cell (Berne etal., 1998).

Rea et al. (1991) tested over 100 patients who believed they weresensitive to electrical exposure by challenging them to respond to2900 nT at the floor, 350 nT at the level of the chair seat, of frequen-cies ranging from 0.1 Hz to 5 kHz. He found that 16% of the patientsresponded to 100 percent of the test signals which were repeatedrandomly 3 times.

Many signs and symptoms of such human common complaintsas chronic fatigue syndrome, fibromyalgia, and myofascial pain syn-drome may be caused by toxicities such as electrohypersensitivity(Genuis and Lipp, 2011). Electrical exposure often proliferates andexacerbates multiple chemical sensitivities (Scarfi, 2008).

The Electropathic Stress Syndrome is a manifestation of the GeneralAdaptation Syndrome developed by Dr. Hans Selye, M.D. In every as-pect of stress studied the uniform systems were (1) enlargement ofthe adrenal cortex with histological signs of hyperactivity, (2) thymicand lymphatic involution with changes in the blood picture, and(3) gastrointestinal ulcers, usually accompanied by other manifesta-tions of shock (Selye, 1950, 1951; Turner, 1955). Selye and colleaguesat the University of Montreal, Quebec, Canada, published some 1500reports and 27 books describing a lifetime of research defining the ef-fects of stress on animals and man.

EMF proliferates and exacerbates: allergies, asthma, Alzheimer'sdisease, brain tumors, strokes, CNS cancer, leukemia, breast, ovarian,prostate and testicular cancer; heart arrhythmia–atrial fibrillation.EMF interrupts communication between cells, enzyme action, ATP en-ergy transfer, and neuroendocrine control of the autonomic nervoussystem, homeostasis; interrupts immune defense, reproduction, neu-roendocrine response of adrenals, thyroids, gonads, and other glandsas noted above and in the references (Genuis and Lipp, 2011;Johansson, 2006, 2007; Havas, 2006; Havas and Olstad, 2008; Rea etal., 1991; Marino and Ray, 1986; Cherry, 2001; Taylor, 2009; Hillman,2009a, 2009b; Milham, 2010; Sage and Carpenter, 2012; Li et al., 2011).

5. Conclusions

Dairy cows were sensitive to earth currents associated with tran-sients recorded in neutral-to-ground circuit outlets, from the floor inmilking stalls, and in barn yards of twelve farms studied. Ground volt-age as low as 10 mV p-p adversely affected milk production. Step


potential voltages recorded by oscilloscope from metal plates in thefloor of cow-stalls and in watering tanks were non-sinusoidal distor-tions of the 60-Hz waveform having frequencies ranging from the1st through the 42nd harmonic, and up to 30 MHzwith numerous im-pulses in overload, exceeding the capacity of the oscilloscope. Thequality of electric power on farms and power lines affecting cattlewas inferior to the 60-Hz steady-state sinusoidal current describedin “stray voltage” laboratory reports (Appleman and Gustafson,1985; Lefcourt and Akers, 1982; Stray Voltage Symposium, 1984;USDA, 1991).

Cow's behavior, health, and milk production were negatively re-sponsive to harmonic distortions of step-potential voltage, suggestingthat utility compliance with IEEE standards on dairy farms needs tobe addressed. Measures of step potential were recorded by battery-powered oscilloscope while the primary power supply was complete-ly disconnected from each farm. During this period there were nodetectable changes in spikes on waveforms indicating the inferiorpower was from off-farm sources transferred on the neutral wire,uninterrupted in a grounded-Wye distribution system.

Power quality varied greatly from farm to farm and day to day.Milkproduction responses to changes in power quality varied inverselywith the number of transient events recorded with event recorders,oscilloscope, and power quality meters. Harmonics often gave betterestimates of electrical effects on milk production than voltage per se.

Peak-to-peak values were correlated with changes in milk produc-tion which permitted measuring the partial effects of independentvariables of electrical currents on milk production using multiple re-gression analysis in the present research.

Use of a 500-Ohm resistor in the volt meter test circuit, for powerquality, effectively eliminates from consideration the electrical mea-sures that were found to be harmful in this study andmay givemislead-ing or unreliable information to investigators and herd owners. Studiesof the effects of various electrical frequencies and harmonics on animalsand humans and the physiological processes affecting behavior, health,reproduction and productivity deserve further attention.

Because power company employees and public service commis-sions are unable to find transients and harmonics in stray voltage, itwould be advisable for all of them as well as professors of electricityto read Barry Kennedy's Power Quality Primer (Kennedy, 2000).

IEEE-SA, Standards Association Marketing Manager Shuang Yuannounced, 25 April 2011, that the IEEE Standards Board approvednew projects that will limit the injection of harmonic frequenciesinto the public electric transmission system. The release said further:“Harmonic pollution is a growing problem caused by the widespreaduse of power supplies and other non-linear loads. It can result inpower loss and equipment damage and it may also be related to envi-ronmental safety issues. Both standards will address harmonic injec-tion in 60-Hz and 120-V/240-V systems such as those in use in theUnited States, Canada, and other regions of the world. Both standardswill also use the IEC SC77A and IEC 61000-3-12 standards as seeddocuments.” The IEEE Standards Association should include harmoniccurrent effects on human and animal health as well as effects on elec-trical equipment.

Acknowledgments

Special appreciation is expressed to Jeff Goeke-Smith, NetworkPlanner and Security Engineer, Michigan State University, who assistedhis father, Charles Goeke, and Dave Stetzer in recording and analyzingdata, and to Dr. Edward Rothwell and Dr. Kun Mu Chen, Professorsand Electrical Engineers, Michigan State University, for their helpfulsuggestions and for confirming electricalmeasurements in East Lansing,MI, homes. Also, Mary Hillman, for preparing and editing the manu-script, and all the farm families who shared their technical, financial,and personal information for the benefit of neighbors and advancementof public knowledge.

References

Alila HW, Hansel W. Induction of lymphopenia causes luteal dysfunction in cattle. BiolReprod 1984;31:671–8.

Aneshansley DJ, Czarniecki CS. Complex electrical impedance of cows: measure-ment and significance. Presentation paper no. 90-3509. St. Joseph, MI: ASAE;1990.

Aneshansley DJ, et al. Cow sensitivity to electricity during milking. J Dairy Sci 1992;75:2733–41.

Aneshansley DJ, et al. Holstein cow impedance from muzzle to front, rear, and allhooves. Presentation Paper No. 95-3621. St. Joseph, MI: ASAE; 1995.

Appleman RD, Gustafson RJ. Source of stray voltage and effect on cow health and per-formance. J Dairy Sci 1985;68:1554–7.

Bawin SM, et al. Effects of modulated VHF fields on the central nervous system. Ann NYAcad Sci 1975;247:74–81. http://dx.doi.org/10.1111/j.1749-6632.1975.tb35984.x.

Behr M. Stray Voltage Research Fraud. 3rd ed. Monograph, 1997, [Self-published, MPSCCase No. U-11684, Exhibit No: C (MB-3), Michael Behr, Jan 2002].

Berne RM, et al. Physiology. 5th edition. Philadelphia, PA: Mosby, Elsevier, Inc.; 1998.Blackman CF, et al. Effects of ELF (1–120 Hz) and modulated (50 Hz) RF fields on the

efflux of calcium ions from brain tissue in vitro. Bioelectromagnetics 1985;6:1-11.

Brennan TM, Gustafson RJ. Behavioral study of dairy cow sensitivity to short AC currents.Paper No. NRC-86-202. ASAE-NRC-CSAE Meeting, Winnipeg, Manitoba, Sept. 26–27;1986.

Brown MR, et al. STRESS; Neurobiology and Neuroendocrinology. New York, Basel,Hong Kong: Marcel Dekker; 1991 [Opioids, pp. 340+].

Buchner K, Eger H. Changes of clinically important neurotransmitters under the influ-ence ofmodulated RF fields— A long-term study under real-life conditions. UmweltMed-Ges 2011;23(1):44–57. [Original in German].

Burch JA, et al. Melatonin metabolite levels in workers exposed to 60 Hz magnetic fields:work in substations and with 3-phase conductors. J Occup Environ Med 2000;42(2):143–50.

Burchard JF, et al. Effects of electromagnetic fields on the levels of biogenic amine me-tabolites, quinolinic acid, and β-endorphin in the cerebrospinal fluid of dairy cows.Neurochem Res 1998;23(12):1527–31.

Burchard JF, et al. Macro and trace element concentrations in blood plasma and cere-brospinal fluid of dairy cows exposed to electric and electromagnetic fields.Bioelectromagnetics 1999;20:358–64.

Burchard JF, et al. Effect of 10 kV/m and 30 μT, 60 Hz electric and magnetic fields onmilk production and feed intake in nonpregnant dairy cattle. Bioelectromagnetics2003;24:557–62.

Burchard JF, et al. Biological effects of electric and magnetic fields on productivity ofdairy cows. J Dairy Sci 1996;79(9):1549–54.

Chen Kun-Mu, et al. Quantification of interaction between ELF-LF electric fields andhuman bodies. IEEE Trans Biomed Eng 1986;33(8):746–56. [Aug].

ChenG, UphamBL, SunW, ChangCC, Rothwell EJ, ChenKM, et al. Effects of electromagneticfield exposure on chemically induced differentiation of Friend erythroleukemia cells.Environ Health Perspect 2000;108(10):967–72. [Oct].

Cherry N. Evidence that electromagnetic fields from high voltage powerlines and inbuildings, are hazardous to human health, especially to young children. NZ: HumanSciences Division, Lincoln University; 2001 [[email protected]].

Craine LB. Effects of distribution system ground voltages appearing on domestic waterlines. Paper No. 69-814. St. Joseph, MI: ASAE; 1969.

Craine LB. Liability for neutral-to-earth voltage on farms. Presentation Paper No.82-3510.St. Joseph, MI: ASAE; 1982.

Craine LB, et al. Electric potentials and domestic water supplies. Agricultural EngineeringSt. Joseph, MI: ASAE; 1970.

Crumpton MJ, Collins AR. Are environmental electromagnetic fields genotoxic? Hottopics. DNA Repair 2004;3:1385–7. [Science Direct© ELSEVIER].

Dahmen N, Ghezel-Ahmadi D, Engel A. Blood laboratory findings in patients sufferingfrom self-perceived electromagnetic hypersensitivity (EHS). Bioelectromagnetics2009;30:299–306.

Dasho DM, Cook M, Reines R, Reinemann DJ. Characteristics of cow contact voltagetransients. ASAE presentation paper no. 943602, Atlanta, GA; 1994 [MPSC Case #U-11684, Exhibit R- (DJR-26), Witness: Douglas J Reinemann, Feb 20, 2001].

De Andrea P. Understanding how the power quality performance of an electric utilitydelivery system affects customer equipment. Paper No. 99-306, July 18–21, 1999ASAE meeting presentation, Toronto, Ontario, Canada. New York State Electricand Gas Corporation; 1999.

Del Giudice E, Giuliani L. Coherence in water and the kT problem in living matter. Eur. J.Oncology Library, vol. 5. “Bernado Ramazzini,” Bologna, Italy: Nat. Institute for theStudy and Control of Cancer and Environmental Diseases; 2010.

Delgado J, et al. Embryological changes induced by weak, extremely low frequencyelectromagnetic fields. J Anat 1982;134:533–51.

Diem E, Schwarz C, et al. Non-thermal DNA breakage by mobile phone radiation(1800 MHz) in human fibroblast and in GFSH-R17 rat granulose cells in vitro.Mutat Res Genet Toxicol Environ Mutagen 2005;583:178–83.

Eskander EF, et al. How does long term exposure to base stations and mobile phonesaffect human hormone profiles? Clin Biochem 2011. http://dx.doi.org/10.1016/j.clinbiochem.2011.11.006.

Genuis SJ, Lipp C. Review: electromagnetic hypersensitivity: fact or fiction? Universityof Alberta, Canada. Science Direct – Science of the Total Environment. Elsevier;2011.

Gorewit RC, Scott NR, Czarniecki CS. Responses of dairy cows to alternating electricalcurrent administered semi-randomly in a non-avoidance environment. J Dairy Sci1984a;68:718–25.

http://dx.doi.org/10.1111/j.1749-6632.1975.tb35984.x


http://dx.doi.org/10.1016/j.clinbiochem.2011.11.006

http://dx.doi.org/10.1016/j.clinbiochem.2011.11.006


Gorewit RC, Drenkard DV, Scott NR. Physiological effects of electrical current on dairycows. Proceedings of the national stray voltage symposium, Oct. 10–12, 1984,Syracuse, NY. ASAE Pub.St. Joseph, MI: ASAE; 1984b.

Gorewit RC, et al. Effects of voltage on cows over a complete lactation: 1. Milk yield andcomposition 2. Health and reproduction. J Dairy Sci 1992;75:2719–32.

Graham MH. Mitigation of electrical pollution in the home. Memorandum No. UCB/ERLMO2/8, 19 Apr 2002, Electronics Research Laboratory, College of Engineering, U ofCA, Berkeley, CA.

Graham, MH. A microsurge meter for electrical pollution research. Memorandum No.UCB/ERL MO3/3, 19 Feb 2003, Electronics Research Laboratory, College of Engi-neering, U of CA, Berkeley, CA.

Graham MH. A method for measuring stray current in ambulatory cows. Paper No.963084, International Meeting ASAE, St. Joseph, MI; 2006.

Gruesenmeyer DC. Measurement and interpretation of neutral-to-earth voltage andtransient voltage surges in milking parlors. J Dairy Sci 1980;63:1-119. [Suppl.].

Gustafson RJ, Cloud HA. Circuit analysis of stray voltage sources and solutions.Transactions of the ASAESt. Joseph, MI: Electric Power and Processing Division;1982.

Gustafson RJ, Albertson VD, Kinney LL. Agricultural electronic equipment operating inan electric transmission line environment. Paper No. NCR 79-204, ASAE AnnualMeeting, St. Paul, MN; 1979.

Hartsell D, Dahlberg D, Lusty D, Scott R. The effects of ground currents on dairy cows: acase study. The Bovine Practitioner 1994:71–8. [September].

Harvey WR. User's guide for LSMLMW and MIXMDL, P-C 2 Version mixed modelleast-squares maximum likelihood computer program; 1990 [OH State University,Columbus, OH, and ARS-USDA].

Havas M. Electromagnetic hypersensitivity: biological effects of dirty electricity withemphasis on diabetes and multiple schlerosis. Electromagn Biol Med 2006;25(4):259–68.

Havas M, Olstad A. Power quality affects teacher wellbeing and student behavior inthree Minnesota schools. Sci Total Environ 2008;402(2–3):157–62.

HillmanD. Effects of electrical shockon cattle. National Electrical Code Internet Connection,Copyright© 2002 Mike Holt Enterprises, Inc, 1-888-NEC-CODE (1-888-632-2633).

Hillman D. EMF in homes, schools, and workplaces: dangers of dirty electricity onhumans and animals (2 DVD presentations). 25th International symposium onman and his environment in health and disease; 2007. [Am Environ Health Foun-dation, Dallas, TX, and U of North Texas, Fort Worth, TX].

Hillman D. Tensen Family Farm – Investigative Report. unpublished report to TensenFamily, Ravenna, MI and to MPSC, 2008.

Hillman D. Cardiovascular response to electric and magnetic fields (DVD presentation).27th International symposium on man and his environment in health and disease;2009a. [AmEnvironHealth Foundation, Dallas, TX, andUNorth Texas, FortWorth, TX].

Hillman D. The electropathic stress syndrome: neuroendocrine relationships (DVD pre-sentation). 27th International symposium on man and his environment in healthand disease; 2009b. [Am Environ Health Foundation, Dallas, TX, and U North Texas,Fort Worth, TX].

Hillman D. Extraneous electrical current at the John and Carol Szymanski Dairy Farmand damages to the cattle and farm income. Sanilac County, MI. unpublished reportto Szymanskis, Court, and Sanilac County Health Department, 2012.

Hillman D, Stetzer D, Graham M, Goeke CL, Van Horn HH, Wilcox CJ, et al. Relationshipof electric power quality to milk production of dairy herds. ASAE presentationpaper no. 03-3116, LasVegas, NV; 2003a.

Hillman D, Stetzer D, Graham M, Goeke CL, Van Horn HH, Wilcox CJ, et al. Relationshipof electric power quality to milk production of dairy herds. CSAE-CSGR presenta-tion paper no. 03-505, Montreal, Quebec, Ca; 2003b.

Hillman D, Stetzer D, Graham M, Goeke CL, Van Horn HH, Wilcox CJ, et al. Monitoringelectrical power quality: effects on milk production and behavior of dairy cattle.Phoenix, AZ: ADSA; 2003c.

Hillman D, Goeke C, Moser R. Electric and magnetic fields (EMFs) affect on milk pro-duction and behavior of cows: results using shielded-neutral isolation transformer.12th Int. conf. on production diseases in farm animals, Mich. State Univ., Vet Col.,July 2004; 2004. [Video available].

Hillman D, Stetzer D, Hillman L, Smith P. Report of electrical investigation at the JerryMartin Dairy Farm, Sanilac Co, MI. unpublished report to Martin Family and SanilacCounty Health Department, 2011a.

Hillman D, Stetzer D, Hillman L, Smith P. Report of electrical investigation atZimmerman Dairy Farm and Home. Sanilac Co, MI. unpublished report toZimmerman Family and Sanilac County Health Department, 2011b.

Hoben PJ (Secretary), Staehle, Anderson LE, Dziuk HE, Hird D, Liboff AR, et al. Final re-port of the science advisors to the Minnesota Public Utilities Commission; 1998.

IEEE 519. Recommended practices and requirements for harmonic control in electricpower systems. IEEE Industry Applications Society/Power Engineering Society,IEEE Standard 519-1992; 1993. [April 12].

Ivancsits S, et al. Induction of DNA strand breaks by intermittent exposure toextremely-low-frequency electromagnetic fields in human diploid fibroblasts.Mutat Res 2002;519:1-13. [In Crumpton, 2004].

Ivancsits S, et al. Intermittent extremely-low-frequency electromagnetic fields causeDNA damage in a dose-dependent way. Int Arch Occup Environ Health 2003;76:431–6. [In Crumpton, 2004].

Johansson O. Electrohypersensitivity: state-of-the art of a functional impairment.Electromagn Biol Med 2006;25:245–58.

Johansson O. Evidence for EMF effects on the immune system. Stockholm, Sweden: TheExperimental Dermatology Unit, Department of Neuroscience, Karolinska Institute;2007 [In BioInitiative Working Group, Sect. 8, July 2007, www.Bioinitiative.org].

Juutilainen J, et al. Relationship between field strength and abnormal development inchick embryos. Int J Radiat Biol 1987;52:787–93.

Kabuto MH, et al. A case–control study of childhood leukemia and residential power-frequency magnetic fields in Japan. Int J Cancer 2006;119(3):643–50.

Kaune WT, et al. Study of high- and low-current-configuration homes from the 1988Denver childhood cancer study. Bioelectromagnetics 2002;23:177–88.

Kennedy BW. Power quality primer. New York, NY: McGraw-Hill; 2000.Kristula M, et al. Effects of prostaglandins F2α synchronization program in lactating

dairy cattle. J Dairy Sci 1992;75:2713–8.Lanera D, et al. Study of magnetic fields from power-frequency current on water lines.

Bioelectromagnetics 1997;18:307–16.Lefcourt AM, Akers RM. Endocrine response of cows to controlled voltage during

milking. J Dairy Sci 1982;65:2128.Lefcourt AM, et al. Responses of dairy cows to non-avoidable electric shock. Soc

Neurosci Abstr, 8. ; 1982. p. 461.Li J, Heber AJ, Johnson GL. Electric power harmonics at rural substations and farms.

Trans ASABE 1990;33(6):1250.Li DK, et al. A population-based prospective cohort study of personal exposure in mag-

netic fields during pregnancy and the risk of miscarriage. Epidemiology 2002;13(1):9-20.

Li DK, et al. Exposure to magnetic fields and the risk of poor sperm quality. ReprodToxicol 2010;29(1):86–92.

Li DK, et al. Maternal exposure to magnetic fields during pregnancy in relation to therisk of asthma in offspring. Arch Pediatr Adolesc Med 2011;165(10):945–50.http://dx.doi.org/10.1001/archpediatrics.2011.135.

Liboff AR. Geomagnetic cyclotron resonance in living cells. J Biol Phys 1985;9:99-102.Loomis DP, Savitz DA. Mortality from brain cancer and leukemia among electrical

workers. Br J Ind Med 1990;47:633–8.Ludington DC, Pellerin RA, Aneshansley DJ. Transmission of neutral to earth currents in

dairy barns. Paper no. 87-3032. St. Joseph, MI: ASAE; 1987.Maisch D. The Procrustean Approach – Setting exposure standards for telecommunica-

tions frequency electromagnetic radiation. A thesis submitted in fulfillment of therequirements for the award of the degree Dr. of Philosophy from U of Wollongong,Australia, 2010.

Maisch D. Comments and recommendations on the draft report: safety of advancedmetering infrastructure in Victoria; 2012 [May 17, Tasmania, Australia, http://www.emfacts.com/papers/].

Marino AA, Ray J. The electric wilderness. Powerline electromagnetic field and humanhealth. San Francisco Press, ISBN: 0-911302-55-7; 1986 [http://www.ortho.lsumc.edu/Faculty/Marino/EW].

Marino AA, Berger TJ, Austin BF, Becker RO, Hart FX. In vivo bioelectrochemical changesassociated with exposure to extremely low frequency electric fields. Physiol ChemPhys 1977;9:433–41.

Marks TA, Ratke CC, EnglishWO. Stray voltage and developmental, reproductive and othertoxicology problems in dogs, cats, and cows. Vet Hum Toxicol 1995;37(2):163–72.

Mazur GA. Power quality, measurement and troubleshooting. Homewood, IL: Ameri-can Technical Publishers, Inc.; 1999.

McLeod BR, Smith SD, Liboff AR. Potassium and calcium cyclotron resonance curvesand harmonics in diatoms (A. coffeaeformis). J Bioelectr 1987;6:153–68.

Milham SD. Dirty electricity: electrification and the diseases of civilization. New York,Bloomington, Indiana, USA: IUniverse, Inc.; 2010.

Milham SD, Stetzer D. Dirty electricity, chronic stress, neurotransmitters and disease.Electromagnetic Biology and Medicine, in press [Informa Healthcare, Ahead ofPrint. http://informahealthcare.com/doi/pdf/10.3109/15368378.2012.743909].

Minnesota Public Utilities Commission. Research findings and recommendations re-garding claims of possible effects of currents in the earth on dairy cow health andmilk production. Final report of the science advisors to the MN PUC, St. Paul, MN,USA; 1998.

Moh'd-Ali AA, et al. Effects of extremely low-frequency magnetic fields on fertility ofadult male and female rats. Bioelectromagnetics 2001;22:340–4.

MPSC E-Docket Case # U-16129. Testimony of Hillman, D – STATE OF MICHIGANBEFORE THE MICHIGAN PUBLIC SERVICE COMMISSION In the matter of the com-plaint of Tensen Family Farms, LLC, 2011; #0101, pp. 134–142. http://efile.mpsc.state.mi.us/efile/docs/16129/0101.pdf.

MPSC E-Docket Case # U-17000. Statement of Jernigan, J, Lamphear, B, Hillman, D.2012 Document No. 0236, http://efile.mpsc.state.mi.us/efile/docs/17000/0236.pdf.

Najdawi IM, et al. Harmonic trend in the NE USA: A preliminary review.Worcester Poly-technic Institute, Worcester, MA, and New England Power Service Co., Westboro,MA. PE-266-PwRD-0-01-1999. (1999, IEEE Transactions on Power Delivery).

Nakata A, et al. Decrease of suppressor inducer (CD4+CD45RA) T lymphocytes and in-crease of serum immunoglobulin G due to perceived job stress in Japanese nuclearelectric power plant workers. J Occup Environ Med 2000;42:143-150.8.

Neutra R, et al. Review of health risks associated with exposure to powerline EMF athome or in the workplace: report to the CPUC. California Department of HealthServices EMF Program. Sacramento, California: Dept of Health Services; 2001.

Nordenson I, et al. Chromosomal abnormalities in lymphocytes of humans exposed topower-frequency fields. Radiat Environ Biophys 1984;23:191.

Nordstrom S, et al. Genetic defects in offspring of power-frequency workers.Bioelectromagnetics 1983;4:91.

Norell RJ, Gustafson RJ, Appleman RD, Overmire JB. Behavioral studies of dairy cattlesensitivity to electric current. Trans ASAE 1983;165(5):1506–11. [EPPD, (2)Paper No. 82-3520, ASAE, St. Joseph, MI].

Phillips JL, Winters WD, Rutledge J. In vitro exposure to electromagnetic fields: changesin tumour cell properties. Int J Radiat Biol 1986;49:463–9.

Polk C. Cows, ground surface potentials and earth resistivity. Bioelectromagnetics 2001;22:7-18.

Przekop F, et al. The effect of prolonged stress on the oestrous cycles and prolactic se-cretion in sheep. Anim Reprod Sci 1984;7(4).

http://www.Bioinitiative.org

http://dx.doi.org/10.1001/archpediatrics.2011.135

http://www.emfacts.com/papers/

http://www.emfacts.com/papers/

http://www.ortho.lsumc.edu/Faculty/Marino/EW

http://www.ortho.lsumc.edu/Faculty/Marino/EW

http://informahealthcare.com/doi/pdf/10.3109/15368378.2012.743909

http://efile.mpsc.state.mi.us/efile/docs/16129/0101.pdf




Rea WJ, et al. Electromagnetic field sensitivity. Bioelectricity 1991;10(1 & 2):241–56.Reinemann, DJ, et al. Dairy Cow Response to Electrical Environment. Final Report: Part

III. Immune function response to low-level electrical current exposure. Submittedto the Minnesota Public Utilities Commission, June 30, 1999.

Sage C, Carpenter DO, editors. Bioinitiative 2012—A rationale for biologically-based expo-sure standards for low-intensity electromagnetic radiation; 2012. http://bioinitiative.org/.

SAS, Inc. User's guide: statistical version. 5th ed. Cary, NC: SAS Institute; 1985.Scarfi MR. Biological effects induced by combined exposures to electromagnetic fields

and chemical or physical agents: a review. Naples, Italy: ICEMB AT CNR — Institutefor Electromagnetic Sensing of Environment; 2008.

Selye H. Stress and the general-adaptation-syndrome. Br Med J 1950;1:1362.Selye H. The influence of STH, ACTH, and cortisone upon resistance to infection. Can

Med Assoc J 1951;64:489–94.Shaw DW, Britt JH. In vivo oxytocin release frommicrodialyzed bovine corpora lutea dur-

ing spontaneous and prostaglandin induced regression. Biol Reprod 2000;62:726–30.Smith AA. Coupling of external electromagnetic fields to transmission lines. 2nd ed. Inter-

ference Control Technologies, Inc, Library of Congress Cataloging in Publication Data;1989.

Stelletta C, et al. Effects of exposure to extremely low frequency electro-magnetic fieldson circadian rhythm and distribution of some leukocyte differentiation antigens indairy cows. Biomed Environ Sci 2007;20:164–70.

Stetzer D. Effects of low level non-linear voltage applied to cows. Videotape of cattleand horse movements in relation to oscilloscope record of transients. Blair, WI,USA: Stetzer Electric, Inc.; 1999.

Stetzer D. Beyond Coincidence – The Perils of Electric Pollution.© 2001Video. StetzerElectric, Inc., Blair, WI. Part 1 - http://www.youtube.com/watch?v=FU7h9pZ0REsand Part 2, http://www.youtube.com/watch?v=OiGHk48mivA.

Stoebel DP, Moberg G. Repeated acute stress during the follicular phase and leutenizinghormone surge of dairy heifers. J Dairy Sci 1982;65:92–6.

Stray Voltage Symposium. Proceedings of the national stray voltage symposium. Syracuse,NY. St. Joseph, MI: ASAE; 1984.

Taylor JB. My stroke of insight — A brain scientist's personal journey. New York, NY:Plume Printing, Penguin Group; 2009.

Thomas TL, et al. Brain tumor mortality among men with electrical and electronic jobs.A case control study. J Natl Cancer Inst 1987;79(2):233–8.

Tran TQ, et al. Electric shock and elevated EMF levels due to triplen harmonics. IEEETrans Power Deliv 1996;11(2). [April].

Turner D. General endocrinology. 2nd ed. Philadelphia, PA: W. B. Saunders Co.; 1955.USDA-ARS Publication 696. Effects of electrical voltage/current on farm animals: how to

detect and remedy problems.Washington, DC: U.S. Government Printing Office; 1991.Van Wijngaarden E, et al. Exposure to electromagnetic fields and suicide rate among

utility workers: a nested case–control study. Occup Environ Med 2000;57:258–63.Wertheimer N, Leeper ED. Electric wiring configurations and childhood cancer. Am J

Epidemiol 1979;109:273–84.Wertheimer N, Leeper ED. Adult cancer related to electrical wires near the home. Int J

Epidemiol 1982;11:345-355.10.Wojczak L, Romanowski S. Simple model of intermembrane communication by means

of collective excitations modified by an electric field. Bioelectrochem Bioenerg1996;41:47–51.

Zdrojewski J, Davidson JN. A review of the problems associated with stray voltage indairy herds. Bovine Pract 1981;16:54–7.

Zhadin M. On the mechanism of extremely weak magnetic field action on aqueous so-lution of amino acid. Moscow, RussiaEuropean J Oncology. Library, vol. 5. ; 2010.[Bologna, Italy. Also: Biophysics 1996; 41:4:843-860].

http://bioinitiative.org/

http://bioinitiative.org/

http://www.youtube.com/watch?v=FU7h9pZ0REs

http://www.youtube.com/watch?v=OiGHk48mivA

Science of the Total Environment - StopUMTS productie en vuile... · 2018. 5. 24. · 42-Hz square...

Documents

Transcript of Science of the Total Environment - StopUMTS productie en vuile... · 2018. 5. 24. · 42-Hz square...